Cai SCS, Allen JC, Lim YL, Tan SH, Tang MBY. Association of Bullous Pemphigoid and Malignant Neoplasms. JAMA Dermatol. 2015;151(6):665-667. doi:10.1001/jamadermatol.2014.5263
Bullous pemphigoid (BP) is the most common autoimmune-mediated subepidermal blistering skin disease and occurs mostly in elderly persons. The association of malignant neoplasms with BP has been controversial.1 This study compared the incidence of cancer in patients with BP with that of the age- and sex-matched general population.
Our study cohort comprised all newly diagnosed patients with BP who were seen at the National Skin Centre in Singapore from April 1, 2004, to December 31, 2009. A diagnosis of BP was based on the presence of at least 3 of 4 criteria: (1) clinical findings consistent with BP; (2) histopathological findings of subepidermal blister; (3) a direct immunofluorescence finding of IgG and/or complement proteins along the dermo-epidermal junction; and (4) indirect immunofluorescence findings of a roof or roof-and-floor pattern or positive anti–BP180-NC16A IgG antibodies.2 Patients with BP were matched with persons in the Singapore Cancer Registry and Death Registry to identify those who had cancer and/or had died. Person-years were accrued from BP diagnosis until death, cancer diagnosis, or December 31, 2011, whichever was earliest. The expected number of cancer cases was calculated as the exposure multiplied by the incidence within each age- and sex-specific category and then summed across all categories for the period 2004-2008 in Singapore. The standardized incidence ratio was calculated as the ratio of observed to expected cancers, and 95% CIs were calculated for the standard incidence ratio based on Poisson distribution. The study was approved by the ethics committee of the National Healthcare Group, Singapore. Patient consent was waived.
A total of 359 patients with BP (mean age, 75.7 years; range, 27.8-100.8 years) were included (Table 1). Of these, 48 (13.4%) had a malignant neoplasm, of which 16 preceded BP diagnosis by more than 5 years (mean, 17.1 years; range, 6.8-26.9 years). Of the 32 recent cancers, 14 occurred within 5 years preceding BP diagnosis. The remaining 18 occurred either in the same year as a BP diagnosis (n = 4) or after BP diagnosis (n = 14). These included 6 lung cancers, 5 skin cancers (nonmelanoma), 3 colorectal cancers, 2 breast cancers, 1 pancreatic cancer, and 1 prostate cancer. The mean follow-up period from BP diagnosis was 2.5 years (range, 0.2-7.4 years); the mean time from BP diagnosis to cancer diagnosis was 0.9 years (range, –1.0 to –4.2 years). The mean time from cancer diagnosis to death was 1.3 years (range, 0-7.4 years), and 7 of the 15 deaths (46.7%) that occurred during follow-up were cancer related.
The expected number of incident cancer diagnoses after BP diagnosis was 14.48, with 14 cancers observed (within 4.2 years) after BP diagnosis, yielding an age- and sex-matched standard incidence ratio of 0.97 (95% CI, 0.53-1.62) (Table 2). There was no association with malignant neoplasms, even when cancers in the concurrent year but preceding BP diagnosis or up to 5 years preceding BP diagnosis were included. No association was observed with any particular type of malignant neoplasm. While studies initially reported an association between BP and cancer,1 subsequent studies4,5 have refuted this association, including a recent British study.6 Although there is a possibility of increased risk of malignant neoplasm in younger patients with BP,7 we did not find this association.
Our study found no significant increase in the incidence of cancer among patients with BP compared with those in the age- and sex-matched general population. Strengths of this study included a relatively large, well-defined cohort, comprehensive patient capture at a national dermatology referral center, complete verification of cancer incidence and mortality status with the national cancer registry, and long-term follow-up. Bias toward greater surveillance for cancer was unlikely because there was no institutional practice to screen patients with BP for malignant neoplasms.
Our data support the findings that BP is not associated with an increased incidence of malignant neoplasms. Thus, cancer screening may not be indicated based on a diagnosis of BP per se, but it should be individualized according to the age, history, and physical evaluation of the patient.
Accepted for Publication: November 27, 2014.
Corresponding Author: Mark Boon Yang Tang, MD, National Skin Centre, 1 Mandalay Rd, Singapore 308205 (firstname.lastname@example.org).
Published Online: February 11, 2015. doi:10.1001/jamadermatol.2014.5263.
Author Contributions: Drs Cai and Allen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Cai, Tan, Tang.
Acquisition, analysis and interpretation of data: All authors.
Drafting of the manuscript: Cai.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Cai, Allen.
Administrative, technical, or material support: Allen, Lim, Tan, Tang.
Study supervision: Lim, Tang.
Conflict of Interest Disclosures: None reported.
Additional Contributions: We thank our colleagues and nurses, especially those in the Immunobullous Clinic, National Skin Centre, for their expert care in treating our patients. We also thank the team from the National Registry of Diseases Office for their invaluable help and support in providing the cancer and mortality data for this study and enabling us to conduct the data analysis at their premises. None received compensation.